Frozen in Time: Sui Generis Label Shields

Courts that apply the Place/Jacobsen analysis describe their legal analysis as a “binary” inquiry. As Judge Moylan, on the Maryland Court of Special Appeals, writing for the majority explained:

174_{See id. at 210 (observing that “[b]ecause Terry involved an exception to the general}

rule requiring probable cause, this Court has been careful to maintain its narrow scope”). In the only other Fourth Amendment case that uses the term sui generis as a discussion point, the dissent used the label to argue for a more narrow interpretation of an earlier case than the one used by the plurality. See United States v. Harris, 403 U.S. 573, 597 (1971) (Harlan, J., dissenting) (protesting the Court’s relaxation of the probable cause standard by its expansive interpretation of Brinegar v. United States, 338 U.S. 160 (1949), and explaining that an expansive reading was not proper because “Brinegar itself was very carefully limited to situations involving the arrest of those driving moving vehicles, . . . a problem that has typically been treated as sui generis by this Court” (internal citation omitted)).

175_{See Gateway Coal Co. v. United Mine Workers, 414 U.S. 368, 376 (1974)}

(disagreeing with the Third Circuit’s assumption regarding the existence of a “public policy disfavoring compulsory arbitration of safety disputes,” which the Third Circuit had viewed as “sui generis”); see also White v. Regester, 412 U.S. 755, 761–62 (1973) (disagreeing with the district court’s suggestion that Abate v. Mundt, 403 U.S. 182 (1971), “in accepting total deviations of 11.9% in a county reapportionment[,] was sui generis”).

176_{Cf. Dunaway, 442 U.S. at 209 n.11, 210 (requiring that the “intrusion must be}

carefully tailored to the rationale justifying it” and observing that the Court had been “careful to maintain [Terry’s] narrow scope”).

866 OREGON LAW REVIEW [Vol. 88, 829

The raison d’etre for treating a dog sniff as a non-search is that the binary nature of its inquiry, “contraband ‘yea’ or ‘nay’?,” precludes the possibility of infringing any expectation of privacy that society objectively considers to be legitimate. If the possession of narcotics in an automobile or a suitcase is illegitimate, so too is the possession of narcotics in a home. It is the criminal nature of the possession itself that takes the activity out from under the protection of the Fourth Amendment, not the place where the possession occurs.177

These courts focus exclusively on the unlawfulness of contraband possession without any consideration of the circumstances under which the contraband is possessed. For example, in Fitzgerald v.

State, the court distinguished Karo178 and Kyllo179 from a canine home-sniff because both Karo and Kyllo involved the tracking of noncontraband items (ether in Karo and excessive heat in Kyllo) once the item became a “detail of the home.”180 The Fitzgerald court pointed out that ether, for example, was “a non-contraband item with many legitimate, as well as illegitimate, uses.”181 The comparison to detection of methyl benzoate molecules in the canine sniff context is unavoidable. As discussed in Part I, scientific research has shown that detection dogs likely alert to the volatile methyl benzoate molecule, not to cocaine itself.182 Methyl benzoate, like ether, has many legitimate uses and, unlike ether, is probably present in the ordinary household.183 This scientific research therefore undermines these courts’ reliance on the lawful/unlawful character of the substance or item being tracked as a distinguishing basis to support canine home-sniffs. Similar to the thermal imager’s detection of excessive heat in Kyllo, detection of methyl benzoate allows police to

infer that illegal contraband is also present. As Kyllo instructs, police

inferencing about the contents of a home that is made possible by

177_{Fitzgerald v. State, 837 A.2d 989, 1030 (Md. Ct. Spec. App. 2003), aff’d, 864 A.2d}

1006 (Md. 2004).

178_{United States v. Karo, 468 U.S. 705 (1984).} 179_{Kyllo v. United States, 533 U.S. 27 (2001).}

180_{Fitzgerald, 837 A.2d at 1036 (quoting Kyllo, 533 U.S. at 38).}

181_{Id. (observing that “[e]ther is not contraband and its mere possession is entirely}

lawful. . . . Thus, Karo is factually distinct from both Place and Jacobsen, where the procedure disclosed only the presence or absence of a contraband item” (quoting United States v. Colyer, 878 F.2d 469, 474 n.5 (D.C. Cir. 1989)) (emphasis omitted) (omission in original)).

182_{See supra notes 32–38 and accompanying text.}

183_{See supra note 38 and accompanying text (discussing methyl benzoate as being}

2009] Has the Fourth Amendment Gone to the Dogs? 867

sense-enhancing technology is a “search” under the Fourth Amendment.

The potential detection of noncontraband information was crucial to one federal court asked to consider a suspicionless canine sniff by an explosives-detection dog at a traffic stop.184 Although the facts in

United States v. Esparza are clearly analogous to those of Caballes,

the judge distinguished the two sniffs on the fact that explosives- detection dogs are “trained to detect ammonium nitrate, a chemical found in [ordinary] household items such as fertilizer and printer cartridges.”185 Therefore, similar to the thermal imager at issue in

Kyllo, the sniff by an explosives-detection dog was capable of

detecting lawful activity and thereby violated the Fourth Amendment.186 For purposes of drug-detection sniffs, on the other hand, the Court’s use of the sui generis descriptor has crystallized understanding of such sniffs to the assumptions of the day in 1983, seemingly making impermissible what would otherwise be a clear analogy to Kyllo’s ban on sense enhancement that might reveal noncontraband information.

As a further thought, reliance on the contraband/noncontraband character of the item being tracked is too simplistic from a legal perspective as well. As four members of the Court recently reminded, “[t]he Fourth Amendment does not seek to protect contraband, yet we have required suppression of contraband seized in an unlawful search.”187 Therefore, courts that focus exclusively on the illegality of the item are missing the point. The Supreme Court requires

184_{United States v. Esparza, No. CR-07-14-S-BLW, 2007 U.S. Dist. LEXIS 66455, at}

*6 (D. Idaho Sept. 7, 2007) (finding a suspicionless sniff of a vehicle by an explosives- detection dog was a “search” because the dog detected both contraband and noncontraband items and because the facts did not raise any “special need” to sniff for explosives based on any imminent danger to national security).

185_{Id. at *6. Although not discussed in the case, the judge’s findings on this issue}

appear to be borne out by the scientific literature concerning explosives-detection sniffs. Explosives-detection dogs “respond to the most-volatile compounds present in an explosive, not necessarily to the explosive species itself.” See Sniffers, supra note 34, at 207 (explaining that when detecting plastic explosives, the dog is not responding to the explosive component RDX, “which has a very low vapor pressure,” but instead “to compounds like cyclohexanone, a solvent used in RDX production”).

186_{Esparza, 2007 U.S. Dist. LEXIS 66455, at *7.}

187_{See Hudson v. Michigan, 547 U.S. 586, 621 (2006) (citing Kyllo v. United States,}

533 U.S. 27, 40 (2001) (Breyer, J., dissenting)); see also United States v. Jeffers, 342 U.S. 48, 53–54 (1951) (explaining that “Congress, in abrogating property rights in [contraband drugs], merely intended to aid in their forfeiture and thereby prevent the spread of the traffic in drugs rather than to abolish the exclusionary rule formulated by the courts in furtherance of the high purposes of the Fourth Amendment”).

868 OREGON LAW REVIEW [Vol. 88, 829

determinations of whether an unlawful search has occurred and whether it is appropriate to apply an exclusionary remedy under the circumstances.188 By categorizing all canine sniffs as permissible, and focusing exclusively on the object of the sniff, these lower courts never analyze the lawfulness of the circumstances of the underlying sniff. Courts have never accepted such a simplistic model. For example, where a canine sniff of a person is contemplated, courts have routinely required suspicion, at least in nonborder situations.189 While the courts in the schoolchild sniff cases focused on the fact that a person, rather than an unattended item, was being sniffed,190 the key point is that these courts analyzed the circumstances under which the canine sniff was performed. In other words, these courts recognized that canine sniffs were not per se outside the boundaries of Fourth Amendment protections. By focusing on the context of the sniff, these courts concluded that the sniff of a schoolchild was too intrusive to be performed without individualized suspicion.191 The school sniff cases demonstrate that considering the circumstances of a canine sniff is nothing new, and serve as a clear indicator that evaluating the intrusiveness of a canine sniff is appropriate in other privacy-sensitive circumstances as well. Refusing to consider the context of the canine sniff, in favor of focusing on the contraband for which the dog is sniffing, is therefore wrong.

188_{See Hudson, 547 U.S. at 599 (finding “knock-and-announce” violation but refusing}

to suppress evidence seized in a search pursuant to a search warrant because imposition of an exclusionary remedy was “unjustified”).

189_{The Courts of Appeals for the Fifth and Ninth Circuits have concluded that a canine}

sniff of a schoolchild is a “search” that required a showing of individualized suspicion.

See B.C. v. Plumas Unified Sch. Dist., 192 F.3d 1260, 1267–68 (9th Cir. 1999); Horton v.

Goose Creek Indep. Sch. Dist., 690 F.2d 470 (5th Cir. 1982). But see Doe v. Renfrow, 631 F.2d 91 (7th Cir. 1980) (finding close contact sniff was not a search). For a further discussion of the issue, United States v. Kelly, 302 F.3d 291, 295 (5th Cir. 2002) (permitting the suspicionless canine sniff of a person at an international border).

190_{See Plumas Unified Sch. Dist., 192 F.3d at 1266 (distinguishing between the canine}

sniff of a person and unattended luggage); Horton, 690 F.2d at 478 (recognizing that “the interest in the integrity of one’s person, and the fourth amendment applies with its fullest vigor against any intrusion on the human body”).

191_{See Plumas Unified Sch. Dist., 192 F.3d at 1266 (agreeing with the Fifth Circuit’s}

analysis); Horton, 690 F.2d at 479 (observing that “[i]ntentional close proximity sniffing of the person is offensive whether the sniffer be canine or human”).

2009] Has the Fourth Amendment Gone to the Dogs? 869

3. “Remoteness” as a Justification for Excluding Canine Sniffs from Fourth Amendment Requirements: Possible Semantic and Temporal Interpretations and Their Impact on Canine Sniff Jurisprudence

To determine whether a canine home-sniff is tied closely enough to the justifications for treating the technique as a nonsearch, it is important to examine the Court’s expectations and assumptions concerning the canine sniff technique. As the Jacobsen Court observed, “[h]ere, as in Place, the likelihood that official conduct of the kind disclosed by the record will actually compromise any legitimate interest in privacy seems much too remote to characterize the testing as a search subject to the Fourth Amendment.”192

On the one hand, this discussion provides clear substantiation of an accuracy-based justification for the police investigative tool at issue. In other words, the Jacobsen Court, in its “remoteness” discussion, could be characterizing the testing involved, both the field testing and the canine sniff described in Place, as being so accurate that the odds of not finding contraband, and therefore instead finding private, noncontraband information in the ensuing search, are “much too remote” to view the police investigative tool as a “search.” This semantic interpretation of “remoteness” is consistent with Place in that Place’s description of detection dogs as sui generis appeared to be based on both the dog’s accuracy and the limited intrusiveness of the sniff itself.193 Additionally, lower courts have explained remoteness in semantic terms as well. For example, in Fitzgerald v.

State, the court viewed the likelihood that a drug-detection dog would

alert on medically prescribed marijuana as too remote to be meaningful for purposes of Place.194

On the other hand, “remoteness” could also be interpreted in a temporal sense. The idea here would be that the eventual search of the person’s now-suspicious item should be severed analytically from

192_{United States v. Jacobsen, 466 U.S. 109, 124 (1984) (emphasis added).}

193_{See United States v. Place, 462 U.S. 696, 707 (1983) (observing that determining}

whether contraband is present through a canine sniff does not require opening the suitcase and implicitly assuming the accuracy of the technique); see also Illinois v. Caballes, 543 U.S. 405, 410 (2005) (Souter, J., dissenting) (observing that classification of the canine sniff technique as “sui generis” was based on the limited intrusiveness of the sniff and its accuracy).

194_{See Fitzgerald v. State, 837 A.2d 989, 1028 (Md. Ct. Spec. App. 2003) (observing}

that “[t]he marijuana in the Place case, for instance, might conceivably have been medically prescribed in a state such as California. The critical holding of the Court, however, was not to be foreclosed by a mere ‘remote’ possibility.”), aff’d, 864 A.2d 1006 (Md. 2004).

870 OREGON LAW REVIEW [Vol. 88, 829

the original sniff or field test that produced the suspicion toward the item. A temporal interpretation of remoteness would therefore allow the Court to disconnect the use of the police investigative tool from the inevitable consequences of that use—the search. By looking at these steps in isolation, the Court could ignore the consequences of a false-positive triggering event. In other words, the exposure of private, noncontraband information as a result of a false-positive canine alert would be “too remote” to reflect back, in some constitutional sense, on the search-triggering investigative tool.

The answer to the “remoteness” question has important implications for the ongoing vitality of the Jacobsen premise. If the semantic interpretation for “remoteness” is the proper one, then changes in our understanding of both the accuracy of drug-sniffing dogs, in general, and societal views on what is “contraband” take center stage. In other words, if drug-detection dogs are not as accurate as once assumed or if lawful citizens increasingly store prescription medications in their homes that detection dogs would interpret to be contraband, then the likelihood that legitimate interests in noncontraband information remaining undisturbed is, in fact, not “remote.”

The Supreme Court may be pressing the temporal, rather than the semantic, view of remoteness, however. As the Caballes Court explained:

Although respondent argues that the error rates, particularly the existence of false positives, call into question the premise that drug- detection dogs alert only to contraband, the record contains no evidence or findings that support his argument. Moreover, respondent does not suggest that an erroneous alert, in and of itself, reveals any legitimate private information, and, in this case, the trial judge found that the dog sniff was sufficiently reliable to establish probable cause to conduct a full-blown search of the trunk.195

It is not clear from this statement whether the majority concluded, in some definitional sense, that a false positive from a reliable detection dog, by itself, was incapable of revealing legitimate private information or, instead, that Caballes had simply failed to make this argument. Significantly, however, the context of the Court’s discussion suggests the former and not the latter. Perhaps, the Court recognizes that an accuracy-based foundation for permissive use of canine sniffs is becoming increasingly shaky. To address the problem of false alerts and how such alerts undermine the canine sniff

2009] Has the Fourth Amendment Gone to the Dogs? 871

technique’s justifications, the Court may be willing to sever the connection between the canine sniff and the ensuing search.

Surgically separating the false-positive sniff (which Caballes claims reveals no “legitimate private information”) from the eventual police rummaging in response to the erroneous alert (which apparently is also not a “search” so long as the dog that gave the false alert was “sufficiently reliable”)196 represents a genuine drift beyond the now-suspect accuracy and limited intrusiveness justifications expressed in Place. Further, it is inconsistent with the Court’s earlier express refusal to sever the search and seizure issues in a case involving contraband drugs.197 The Caballes Court’s surprising statement concerning false positives represents an implicit acknowledgment that it needs to patch the hole in canine sniff jurisprudence that has become evident in the years following Place. “Reliable” drug-detection dogs make plenty of mistakes.198 To suggest that a false-positive alert reveals no private information is an artificial conclusion, if ever there was one, because the alert leads directly and inevitably to police rummaging during which private, noncontraband items are uncovered.199

This Caballes dicta may have a real impact on the home-sniff question. There is no data on the accuracy of drug-detection dogs asked to sniff the exterior of a person’s home.200 The data presently available concern the accuracy of detection dogs that are asked to scent in close proximity to the container suspected of secreting contraband (e.g., luggage, a vehicle, or an interior room).201 It is far from clear that existing data concerning luggage and vehicle searches should be unquestioningly extended to establish “reliability” for canine sniffs of the home. First, the detection dog is not able to gain the same proximity to the contraband item as is typically the case

196_Id.

197_{See United States v. Jeffers, 342 U.S. 48, 52 (1951); see also supra note 100.} 198_{See supra notes 151–52, 155-56 and accompanying text.}

199_{As Justice Souter explained, “[n]or is it significant that Kyllo’s imaging device}

would disclose personal details immediately, whereas they would be revealed only in the further step of opening the enclosed space following the dog’s alert reaction; in practical terms the same values protected by the Fourth Amendment are at stake in each case.”

Caballes, 543 U.S. at 413 n.3 (Souter, J., dissenting).

200_{For a discussion of the fact that canine certification for drug detection is limited to}

testing for drugs hidden in vehicles or indoor, interior rooms, rather than perimeter searches of buildings, see supra notes 29–30 and accompanying text.

872 OREGON LAW REVIEW [Vol. 88, 829

during vehicle or luggage sniffs.202 A dog that is “reliable” for purposes of sniffing luggage in close proximity at an airport may not be as effective in a residential setting.203 Significantly, however, no data exist to allow meaningful review of canine reliability in these newer factual situations.204

Second, home occupants have less control over the people who access their front door and associated curtilage areas. Although there are exceptions,205 the front door is an open curtilage location where homeowners typically anticipate interacting with nonfamily members and others. The overall lack of control over who comes and goes from these curtilage areas creates the risk that a drug-detection dog could alert to contraband waste molecules206 left behind by others—

202_{For a discussion of the “scanning” process that drug-detection canines use to locate}

the scent source of narcotics, see supra notes 43–50 and accompanying text. See also Fredric I. Lederer & Calvin M. Lederer, Admissibility of Evidence Found by Marijuana

Detection Dogs, ARMY LAW., Apr. 1973, at 12, 12 (describing a pattern of properly conducted canine-assisted barracks searches). “While the dog may detect airborne scent and follow it to its source, more likely the dog will have to smell the immediate proximity

of an area to detect marijuana within it.” Id. (second emphasis added).

203_{In fact, the scientific literature, discussed in Part I, reveals that proximity is an}

important consideration in both detecting the drug and properly identifying the scent source. See supra notes 43–50 and accompanying text.

204_{See R v. Kang-Brown, [2008] 1 S.C.R. 456, 2008 1 S.C.C. 18 ¶ 15 (Can.); see also}

supra note 51. As the Canadian Supreme Court observed, little empirical research on the

accuracy of detection dogs exists. Kang-Brown, 2008 1 S.C.C. ¶ 15. One study, conducted in Australia, was reported by the Privacy Ombudsman of New South Wales in 2004. The research revealed that seventy-three percent of those searched on the basis of a positive alert from a drug-detection dog were found not to be in possession of illegal drugs. NEW SOUTH WALES OMBUDSMAN,DISCUSSION PAPER:REVIEW OF THE POLICE POWERS (DRUG DETECTION DOGS)ACT 16 fig.3 (2004) [hereinafter NSWOMBUDSMAN 2004]. While sixty-one percent of the false positives were attributable to the “residual odour” thought to be related to the individual’s admission of use or contact with others who had used drugs; thirty-nine percent of the false positives could not be explained. Id. at 23–24. In 2006, the Privacy Ombudsman issued a new report concerning the use of detection dogs. NEW SOUTH WALES OMBUDSMAN, REVIEW OF THE POLICE POWERS (DRUG DETECTION DOGS)ACT (2006) [hereinafter NSWOMBUDSMAN 2006]. Therein, it was determined that seventy-four percent of those searched did not possess illegal drugs.